Why we shouldn’t close down the skills-knowledge debate

“Our consciences dictate that we must protest against it, that we must oppose it and that we must attempt to alter it…”

Nelson Rohihlala Mandela, October 1963, Rivonia Trial


Two great leaders have been on my mind in the last two weeks: JFK and Nelson Mandela. Mandela died this week at the ripe old age of 95. By contrast, Kennedy was assassinated 50 years ago this November, aged 46. What united them was their vision against oppression, their courage against adversity, and their conscience against injustice. Both have inspired millions globally, as the commemorations show. In learning about their lives and teaching their speeches, I have like so many others felt personally inspired by them.

Living in South African and teaching in a township primary school in 2004, 2006 and 2010, I found Madiba’s words inspiring: ‘There is no greater revelation of a society’s soul than the way it treats its children’. Living for a short time in Berlin in 2009, exploring its history, I found Kennedy’s soaring rhetoric resonating with me. He was passionate about education: “Our progress as a nation can be no swifter than our progress in education. The goal of education is the advancement of knowledge and the dissemination of truth.” Fifty years ago this year, while Mandela was on trial for treason in South Africa with the threat of the death penalty or life incarceration looming, Kennedy was in Berlin in 1963. West Berlin had survived a Soviet blockade but was now surrounded by the Berlin Wall. Listening to his words again this week struck me:

“There are many people in the world who really don’t understand, or say they don’t, what is the great issue between the free world and the Communist world.

Let them come to Berlin.

There are some who say Communism is the wave of the future.

Let them come to Berlin.

There are some who say, in Europe and elsewhere, we can work with the Communists.

Lass’ sie nach Berlin kommen. Let them come to Berlin.”


Hearing these words in the characteristic Bostonian lilt should resonate for us in education this week. The international PISA test has again shown Kennedy’s home state of Massachusetts (mean score 511) leading all other states in the USA (mean score 481), competing with the leading Asian jurisdictions, and well ahead of the UK (491). As Kennedy might have said, let them come to Massachussetts: the state with a core knowledge curriculum.

On False Dichotomies

Many in education dismiss the skills and knowledge debate as a false dichotomy. They lament any such polarising rhetoric and position themselves as between or beyond this debate. Of course, the mediators say, we should teach both.

Sir Michael Barber is one of the mediators: ‘the road to hell in education is paved with false dichotomies… It’s an absurd debate: they go together.’


Alex Quigley, a great Head of English and education blogger takes this position in his post Beyond constructivism versus direct instruction and references the Rowe report:

“The relative utility of direct instruction and constructivist approaches to teaching and learning are neither mutually exclusive nor independent. Both approaches have merit in their own right, provided that students have the basic knowledge and skills (best provided initially by direct instruction) before engagement in ‘rich’ constructivist learning activities.’

Alex’s comments represent the ‘beyond binaries’ way of thinking:

“I get tired of the binary debates that fail to recognise the reality of the classroom is far more nuanced than singular methods or ideologies.”

“I use methods from both perspectives on a daily basis…”

“I worry about polarising the knowledge debate too much because the truth is always more complex”

“I find myself in a middle ground in the instructivist/constructivist debate. I like to think I am one of those relativists who cheats taking one side or another and plucks the best from all sides!”

“I’m frustrated by the binary thinking and polarised, ideology laden debates that place direct instruction in opposition to more collaborative learning methods”

“The vast mass of ordinary teachers are outside of much of these polarising debates, busy getting on with teaching.”


Tom Sherrington, great headteacher and education blogger, also takes this view: ‘I’m just really tired of the lack of nuance here.’ Tom also says, ‘Knowledge transmission is a very limiting and limited view of teaching’ and ‘I can’t imagine teaching a lesson sequence in the style Daisy Christodoulou suggests’.

Headteacher Stephen Tierney also shares this perspective: ‘The debate about teacher or pupil-led teaching has to stop, for me it is “wrong jungle” thinking.’


Wrong jungle?

In this blogpost I want to push back on this imperative to get out of the jungle and stop the debate. I know that as moderates and mediators, Alex, Tom and Stephen are open-minded and willing to listen to challenge, as I am: I am prepared for pushback on this, but this is my challenge to them.


Why the knowledge and skills debate is worthwhile

Here are five reasons why we shouldn’t forestall and close down the debate:

  • Evidence-based effectiveness
  • Strategy, choice and opportunity cost
  • Compromise 
  • Ideology: a tale of two icebergs
  • Complacency 

Evidence-based effectiveness

There is now a great deal of statistical, scientific and empirical evidence to show that knowledge and instruction is highly effective for learning, and it is most effective for our most disadvantaged learners. But there is also a research-practice gap: this evidence base is not making its way into teacher training. As Hattie says:

‘Too often direct teaching is portrayed as bad, while constructivist teaching is considered to be good… This is almost directly opposite to the successful recipe for teaching and learning.’ … ‘Every year I present lectures to teacher education students and find that they are already indoctrinated with the mantra “constructivism good, direct instruction bad”. When I show them the results of these meta-analyses, they are stunned, and they often become angry at having been given an agreed set of truths and commandments against direct instruction.’

Hattie presents this table in his latest book, Visible Learning and the Science of How We Learn, distilled from over 900 meta-analyses. It tells its own story:


The flipside is unevidenced approaches that persist. Hattie and Yates say:

“Within both public and professional domains, fallacious ideas of human learning continue to be promoted despite being contradicted by available scientific opinion and evidence. Many such fallacies are potentially destructive and reliant on false premises.”

When such a vast range of statistical, scientific and empirical research tells such a similar story, it’s no good closing down the debate and averting our eyes of the evidence.

Strategy, choice and opportunity cost

The great strategic thinker Michael Porter said that the essence of strategy is choosing what not to do. Both John Hattie and Dylan Wiliam make the point that almost any intervention in education has some impact, so it’s ludicrous to set the bar at zero. Instead of asking what works, we should focus only on what works best. Teachers’ workloads are already very high, without losing focus and beavering away at strategies like individualised instruction that have such small impact for such high effort.

But as Dylan Wiliam says, ‘In professional settings, it is incredibly hard to stop people doing valuable things in order to give them time to do even more valuable things. They reply, “Are you saying what I am doing is no good?”’

In other words, we run the risk of making good the enemy of the best. David Didau is asking the right question here: “Should we stop doing good things? I remember John Hattie urging this at the London Festival of Education in 2012: ‘we need to stop teachers doing things that are just good, in order to get them doing things that have a greater impact.’ The evidence shows that subject knowledge, explicit instruction, worked examples, deliberate practice and guided feedback have a great impact on learning: so why spend time on discovery, inquiry and individualisation, which have such low impact for novice pupils?

There is a terrible opportunity cost in using less effective strategies, especially for our poorest pupils. Disadvantaged students start secondary school years behind in their literacy and numeracy, and millions of words behind in their vocabulary. Every low-impact strategy used is a missed opportunity to narrow the gap.


Counterintuitively, the middle way is not always the right way. If one pupil thinks 20% of 100 is 20, and one thinks it is 30, the answer *in the spirit of compromise* doesn’t suddenly become 25. Tony Blair’s ‘third way’ didn’t work out very well in Iraq, on light-touch regulation in banking, or on tripling education spending whilst national underachievement persisted. When Mandela’s solution was multiracial democracy, and Boethe’s was racial segregation, the answer was not somewhere in between. When Wilberforce’s solution was abolition, and Clarence’s was the continuation of the triangular slave trade, the answer was not somewhere in between. When Kennedy’s solution for Berlin was decentralised liberal democracy, and Khrushchev’s was enforced, command-and-control, state-centralised communism, the answer was not somewhere in between. If one solution is ‘21st century skills’ and the other is ‘core knowledge instruction’, the answer is not somewhere between or even beyond. History shows us that systems based on fallacious ideas collapse under the weight of their own contradictions: the slave trade, the Berlin Wall, the Iron Curtain, apartheid, the banking boom.


Now, as Obama says, educational inequality is the civil rights issue of our time. The skills-only, content-light regime has failed. The ‘outcomes-based’ skills curriculum failed in South Africa, where it entrenched inequality in education, as its proponent and architect Michael Young now acknowledges. The fallacious idea underpinning it is that skills are transferable. They are not; skills depend on domain-specific knowledge. The skills-based curriculum in England has failed over the last decade to tackle educational inequality: 60% of poorest pupils still don’t secure 5 C’s at GSCE, and 20% of school leavers leave school illiterate and innumerate. To urge us to avoid the debate is to perpetuate the skills regime and let it go unchallenged.

With Mandela and Kennedy, those giants of twentieth century leadership, when confronted by injustice, ‘our consciences dictate that we must protest against it, that we must oppose it and that we must attempt to alter it…’ Educational inequality is no different.

Ideology: a tale of two icebergs

Ultimately, questions of what and how to teach come down to a difference in ideas. Here’s a thought experiment: which of these two best represent your ideas about how learning happens?

Image        Image

Have we got our icebergs the wrong way up?

Most teachers I’ve met, after frequent exposure to Bloom’s taxonomy, with low-level knowledge at the bottom and higher-order skills at the top, say that the image of knowledge as just the tip of the iceberg best fits their idea of education.

For many of us, ideology is what the other guy does; that’s a natural bias. The mediators like Alex, Tom and Stephen would like us to get us beyond ideology altogether. But as Daisy Christodoulou points out in her book, teachers who believe themselves quite exempt from any ideology are usually the slaves of some defunct educationalist.


Are we all the slaves of some defunct educationalist*?

It’s no good claiming to be above or beyond ideology; it’s more a matter of working out which ideology most shapes your ideas. And the centre of gravity for at least a decade in teacher training and the educational establishment has been, and still is constructivism.  A casual glance through the Learning to Teach textbook series for new teachers reveals hundreds of pages devoted to Vygotsky and constructivism and none to Willingham and cognitivism. Yet his zone of proximal development yields precious little insight compared to the precise specificity of insight yielded by the cutting-edge science.


When I talk about cultural knowledge in the English curriculum, mastery assessment and teacher training in instruction, the reply I hear is often dismissive: ‘We already do all of this anyway’; ‘It’s what we’ve always done.’

But very few of those who claim to be doing it all already are acting on the insights from evidence-based research and cognitive science. Very few have replaced the skills-based zombie levels with a mastery assessment system. Very few have replaced weak texts with a high opportunity cost at Key Stage 3 like Cirque du Freak or Stone Cold with Oliver Twist and Jane Eyre. Very few have taken the cognitive insight to heart: All that there is to intelligence is the simple accrual and tuning of many small units of knowledge that in total produce complex cognition. Very few have committed to the cognitive imperative: ‘if nothing has been retained in long-term memory, nothing has been learned.’ Very few have redesigned the curriculum with memory in mind. Very few have put Daisy’s approach into practice: the best way to achieve the aim of complex thinking skills for pupils is not to teach skills; it’s to teach knowledge.


A coherent, sequential knowledge curriculum is still a small, young challenger to the gargantuan orthodoxy of the skills curriculum and the skills assessment system in England. In other words, knowledge is the David to the skills regime’s Goliath.



The point of this blogpost is to challenge those calls to move beyond the knowledge versus skills debate and its polarising binaries. I would urge those who make them to rethink these calls, and those who hear them to resist them. We are yet to excavate the full insight from the rich seam of the knowledge and instruction side of the debate. Cognitive science, international comparisons and statistical meta-analyses are not going to go away, and as they become increasingly sophisticated, we must not preclude thinking through this tension carefully. ‘The ability to hold two opposing ideas in mind at the same time and still retain the ability to function is the test of first rate intelligence,’ according to F. Scott Fitzgerald. Let us keep thinking about the effectiveness of those opposing ideas for education.

I realise that this blogpost will likely be criticised for polarising rhetoric. But rhetoric can be a powerful force for changing an iniquitous and intractable status quo. If Mandela was right, and ‘education is the most powerful force you can use to change the world’, it is changing our ideas that holds the greatest hope for improving education. In my next blogpost, I explore the practical differences between a skills-based lesson and a knowledge-led lesson, based on the distinctive and fundamental differences in the ideas underpinning them.

About Joe Kirby

School leader, education writer, Director of Education and co-founder, Athena Learning Trust, Deputy head and co-founder, Michaela Community School, English teacher
This entry was posted in Education. Bookmark the permalink.

25 Responses to Why we shouldn’t close down the skills-knowledge debate

  1. sputniksteve says:

    Hi Joe, I’m really enjoying your blog.


    Sent from my iPhone


  2. David Didau says:

    I’m gutted you didn’t use my epigram that “Compromise is the refuge of the unprincipled”. Too strong?

  3. saidthemac says:

    I absolutely agree, knowledge is fundamental and much of what trainees are told is absolute pap, but let’s look at the example of teachers. Pedagogical content knowledge is useless without content knowledge, and content knowledge must come first.

    Yet pretty much every blog we all write is centred on pedagogical content knowledge. We all seem to value the “how” enough to spend time composing and sharing our thoughts.

    Of course we trust that our peers subject knowledge is up to scratch, as we have all got our degrees and spent our own school careers making sure that this this is the case, but to deny the moderates (which I would suppose I am) the value of an ‘and’ is equally dangerous as the myths that are perpetuated in teaching courses up and down the country.

    Let me make it clear that I don’t think this is what you are doing Joe, but we must not allow skills to be devalued at the expense of the knowledge/skills dichotomy (which is real).

    I am a music teacher, I deal in know how most of the time. The methods I chose are mostly determined by the content. Knowledge about music IS fundamental, but it is the know how I.e. knowledge of music that ultimately is the aim.

  4. Joe, interesting stuff as always.
    Could you give me the source of the Dylan Wiliam quote, I’d like to use that elsewhere.

  5. mrlock says:

    mac – what do you think of Tim’s blog on music here: http://amusosmusings.wordpress.com/2013/12/03/is-music-an-academic-subject/

    • saidthemac says:

      I’m really glad Tim has started blogging. Many of the music teachers who tweet/blog are part of the musical futures community, it’s great to hear a voice from the grammar school domain.

      The recent Ofsted report was widely derided for a perceived bias towards notation based methods. I personally believe in sound before symbol, but the case Tim makes for music being an academic subject is something with which I would agree whole heartedly.

      Tim is clear that theory without context cab be a drag though, his last paragraph indicates so.

      Actually, if you ask kids, they are really keen to learn to read music, and time and again when I see the moment kids “get” some theory, they get a real buzz. The context for theory need not always be practical music making, sometimes the academic study of a piece is appropriate, especially in larger scale works.

      That said, I tell my kids they don’t know a symphony until they can hum it from beginning to end.

  6. You raise many interesting points. I wonder about the meta-analysis you have above. How did they measure the effectiveness of the various models? If the measurement was content-based retrieval of factual information, then I imagine that the teacher led methods were more effective and more efficient. Did they measure development of higher level thinking skills? In an age of information overload, where “facts” are available at the click of a button, where what we imagine to be true today isn’t true tomorrow (Pluto isn’t a planet anymore???), it seems like there is more value in the ability to find, evaluate, and synthesize information than to merely memorize it. However, your point that it is nearly impossible to do that without the knowledge base. So the question is, with limited time how do you do both–teach enough content to form a foundation for higher-level thinking, but to also engage students as producers of information rather than merely consumers? Part of the problem with common core (and I’ve taught in a state that instituted similar standards many years ago), is that you are asking everyone to teach to the middle. For some schools, this means dumbing down and regimenting a program that is already effective. For other schools this establishes standards that children are not close to being able to achieve–in other words setting our most vulnerable children up to fail. The federal government’s attempt to make one size fits all results in mediocracy at best.

  7. Pingback: Should we Compromise on Knowledge and Skills? | Ariadne's Thread

  8. Knowledge is vital- the most important thing, so I broadly agree with the thrust of your argument. But, and it’s a big but, the analogies you use to illustrate your point about not compromising don’t work- you’re not comparing like with like. These are absolutes- to be even more extreme female circumcision should never be compromised on, but this is hardly the same as saying sometimes we need to teach skills, yet it seems (unless I’ve misunderstood) to be the same logic.

    To illustrate my point, I might want students to make a judgement about Mandela being a terrorist or freedom fighter. If they’re going to do it, they need to know a lot of stuff first. However, they’re also going to need to need some help to do that- otherwise their judgements will be trite and not make use of evidence as they ought to. It’s a bit like cooking a Michelin starred plate of food. The ingredients and the amounts are vital, but without the correct techniques I’m not going to get the end product I probably should.

    Subject skills are important then (not generic ones, though there is some cross over with some subjects). I think what we have is not a false dichotomy, but rather the wrong dichotomy. It’s knowledge and skills- knowledge more important but *subject* skills pretty bloody important too.

  9. bt0558 says:

    Fascinating blogpost.

    A couple of questions…..

    “They are not; skills depend on domain-specific knowledge”

    I am interested to know why you would think for one second that because the effective application of skills in a particular situation will depend upon domain-specific (or some other specific) knowledge, that skills are not transferable.

    “Most teachers I’ve met, after frequent exposure to Bloom’s taxonomy, with low-level knowledge at the bottom and higher-order skills at the top, say that the image of knowledge as just the tip of the iceberg best fits their idea of education.”

    Why would you present the issue as two icebergs, one with knowledge at the top and one with skills at the top. An iceberg has most of it’s mass underwater. What is the implication of this for your model.Bloom did not claim that knowledge was at the bottom of the model as it was least important in fact the exact opposite. Knowledge is presented as the basic fodder for the higher order skills and without knowledge, one could not use the higher order skills. Knowledge must come first. This idea was extended by the Anderson and Krathwohl model which addressed some of the issues with Bloom’s, issues that Bloom identified himself.

    The A&K (Bloom) model suggests that remembering facts is a simpler cognitive process than analysis as analysis requires one to remember stuff and then do something with it for analysis. It is not suggested that kinowledge is less important at all.

    Some problems require almost no knowledge and the use of thinking skills. Other tasks require a lot more knowledge and less thinking skills. The model is often drawn as a series of levels as a rectangle and not as a triangle.

    Always good to think about this one. Thanks for the post.

  10. Hi
    I enjoyed reading your blog. I always find your work interesting and thought provoking. Thank you. I am wondering about your comment though: “The fallacious idea underpinning it is that skills are transferable. They are not; skills depend on domain-specific knowledge. “.Are you saying that if I have learnt for example, to analyse by being specifically taught this…an idea in science, that I can then not transfer my knowledge about and therefore my ability to analyse an idea…say in history? If you are saying that then I disagree. My experience with students is that if they learn about for example, what analyse means and practice doing this and use specific tools designed to help with this, that you support them in their learning of how to analyse, work through with them examples of analysis etc then when i ask them to analyse something else they are more able to do this because they have some understanding generically of what analysing entails. What is your experience of this, or have I misunderstood what you are saying, or not understood eg what a domain is? Very interested in your thoughts on this.

  11. I’m looking forward to the next post, but this one was a bit whiny, in my opinion. Quotes from Mandela and JFK? Not really necessary, I think (pushed the blog over the top). Yep, the next one should be more engaging and interesting.

  12. KB says:

    Absolutely! Never compromise on what really matters.

  13. moriarity says:

    Interesting post and one I shared with both colleagues and governors at my school.

    As an aside the the main purpose of the post I do think the referencing of Nelson Mandela, JFK etc. leaves little chance for those at the other side of the debate to take up a position – a little like invoking Godwin’s Law…

    John Hattie also wrote the following:

    “…there is a ‘practice’ of teaching. The word practice, and not science, is deliberately chosen because there is no fixed recipe for ensuring teaching has the maximum possible effect on student learning, and no set of principles that apply to learning for all students. But there are practices we know are effective and those that are not.” (p.5, Visible Learning for Teachers, 2012)

    Hattie also asks teachers to ‘know thy impact’. Later, in chapter 8 of the same book he talks about the ‘facilitative relationship’ (p.157), meaning the essence of a student-centred teacher. I only mention this because the table used above to compare effectiveness of strategies uses the labels teacher/facilitator, which I think is slightly misleading. It might seem like splitting hairs, but if we are to develop as a profession we need to really think about the professional language we use to further the debate. (I have the same concern about the way we use test/assessment in discussions – at my school we decided to define each clearly so as not to confuse the situation as test is a loaded term that carries deep baggage, particularly in the particular culture of the host nation.)

    I have been privileged enough to work in a variety of educational systems, ranging from English NC to IPC, PYP, MYP, IGCSE and IB Diploma. In all those systems I have seen terrible teaching/learning and fantastic teaching/learning. I suspect that if I put those teachers who were most effective in a room and asked them to discuss teaching and learning there would be little between them in terms of approach. While PYP may well be a constructivist based curriculum the best teachers recognise their role is essential and model a gradual release of response through exceptionally careful thought, planning, assessment and observation. What they did within their own classrooms was follow Hattie’s edict. I could say the same about the best teachers I have seen working with a NC environment.

  14. karicketts says:

    Love reading this blog and the ideas explored even though I’m not a sec school teacher – so many thanks. I work with adults to improve workplace skills (mostly journalists and other media professionals) and I’m keen to find out if this debate of skills v knowledge is relevant to adult learning. Is there anybody else out there who is stimulated by the thinking expressed in these blogs but who is not a secondary school teacher?

  15. Pedro says:

    Reblogged this on From experience to meaning… and commented:
    This blogpost reflects some of the elements I already posted earlier here and surely is more than worth reading!

  16. Pingback: How best to teach: knowledge-led or skills-led lessons? | Pragmatic Education

  17. The problem with the icebergs isn’t that they are the wrong way up, but that the iceberg is the wrong metaphor. The iceberg might be a popular analogy for the way individual teachers organise their own understanding of knowledge and skills, but it isn’t necessarily applicable to knowledge and skills in general. I suspect that many of the problems encountered in the education skills-knowledge debate have arisen because the starting point has been the way skills and knowledge have been construed within the education system, rather than with what is known about skills and knowledge in general.

    Imposing either/or categories on anything that occurs naturally (male/female, healthy/ill, intelligent/not intelligent) usually leads to trouble, because nature is complicated. Knowledge cannot realistically be separated from skills, nor is the deep structure of knowledge clearly distinct from analytical skills. You could draw a distinction between knowledge and skills if you wanted to; e.g. knowledge is a body of information and a skill is what you do with that information. But I’m not sure it’s an especially helpful distinction because knowledge and skills are interdependent.

    As I see it, the problem with a skills-based curriculum is not that skills are not transferable, nor that they are domain-dependent because the degree to which skills are transferable and the degree to which they are domain-dependent varies depending on the skill and the knowledge domain involved. What seems to be wrong with the skills-based curriculum is that skills are seen as free-standing and as independent of knowledge, which is meaningless. A skill can’t exist without knowledge and you can’t acquire knowledge without the skill to do so – even if that skill is simply understanding language.

    For example, my ability to gather new information, record it accurately, integrate it into my existing body of knowledge and recall it coherently is a skill that can be applied to all domains, even those I’m unfamiliar with. It has high transferability but low domain-dependence. By contrast, my ability to use laboratory glassware effectively and safely is a skill that has a more limited range of applications. Its transferability and domain-dependence are generally lower, but it has high transferability in relation to cookery. (I’d draw a two-by-two matrix but it wouldn’t work in a comment.)

    One problem in the skills-knowledge debate is that the education system (in England at least) has long seen skills and knowledge as distinct rather than interdependent. Another problem is that teachers often have little understanding of the deep structure of knowledge domains other than the one they specialised in. This is especially problematic for primary teachers, who need to be familiar with a wide range of domains. And I’m not pointing the finger: I well remember the moment 35 years ago, that another 3rd year undergrad and I realised, after a discussion she’d had with her tutor, that the scientific method was a knowledge-domain with a deep structure that neither of us had ever been told about. In fact, despite knowing a lot about science as a domain, what we’d consistently been told is that we *didn’t* need to know about the deep structure of the scientific method. I then did 9 months training which was supposed to equip me adequately for primary teaching. I don’t think it did.

  18. Pingback: Books, bloggers & metablogs: The Blogosphere in 2013 | Pragmatic Education

  19. Pingback: It’s not skills – it’s know-how | Teaching: Leading Learning

  20. Rory says:

    I use a very simple model, all kids must learn 11 basic skills – then they have a chance at bright futures. They are reading, writing, listening, speaking, information organization, critical thinking, study skills and success with math. I also emphasize learning how to strait on time, stay on task and complete assignments.


  21. Pingback: A guide to this blog | Pragmatic Education

  22. Pingback: The Blogosphere in 2016: Roaring Tigers, Hidden Dragons | Pragmatic Education

  23. Pingback: Articles | Joe Kirby

Leave a Reply